Choosing trip rates in a Transport Assessment

Trip generation in a Transport Assessment is decided before any junction is modelled, and once the rates are agreed the rest of the work is arithmetic. Pick the wrong comparator sites and you either over-design the mitigation and pay for a signal scheme nobody needed, or under-state the impact and hand the highway authority a reason to object. The selection method matters more than the number it produces, because a rate you can justify survives scrutiny and a rate you cannot does not.

Agree the approach with the highway authority in writing before the surveys are drawn down. A scoping note that sets out the database category, the filters and the assessment years costs an hour and removes the largest single source of rework.

Why trip rate selection decides the assessment

Everything downstream inherits the rate. Distribution, assignment, junction capacity, parking demand, servicing, the travel plan targets and the section 278 works all scale from it. A 20 per cent difference in the arrival rate can move a priority junction from comfortable to over capacity in the design year.

Highway officers know this, which is why the trip generation chapter attracts more comment than the modelling. They are looking for three things: that the comparator sites resemble your site, that the filtering was not tuned to produce a convenient answer, and that the selection is reported fully enough for them to repeat it.

The database most commonly used in England is TRICS, and its good practice guidance expects the selection process to be documented rather than summarised. Treat that as the standard the chapter is written to.

Building a defensible site selection

Work outwards from the land use category, not from the sample size. Get the category wrong and no amount of filtering rescues the result: privately owned houses behave differently from affordable rented flats, and a retirement scheme behaves differently again.

A workable sequence:

1. Select the land use category that matches the proposal, and say why any related category was rejected.
2. Set the range for the size parameter, whether dwellings, bedrooms or gross floor area, around the proposal rather than symmetrically.
3. Apply location and parking filters that reflect the site.
4. Exclude surveys with characteristics that would distort the rate, naming each exclusion and the reason.
5. Report the number of sites remaining and accept a smaller reliable sample over a larger mixed one.
6. Run a sensitivity test with one filter relaxed, so the officer can see how stable the rate is.

Five to ten comparator sites is a reasonable working target for a residential scheme. Below that, the average becomes hostage to one unusual survey, so quote the range and the standard deviation as well as the mean.

The discipline is the same one described in our note on the baseline chapter of an environmental statement: the credibility comes from showing the selection, not from asserting the conclusion.

Filters that change the answer

Some filters move the rate materially and some barely move it. The ones worth arguing about:

Two filters are frequently misused. Excluding sites with no parking, on the basis that your site has parking, can remove exactly the urban comparators the authority considers relevant. And filtering by survey year to the most recent years alone often leaves too few sites to be meaningful, so the honest move is to include the older surveys and discuss the direction of any bias.

Where the highway authority has its own preferred comparators from previous applications in the area, ask for them at scoping. Officers will often share the rates used on a nearby scheme, and matching them is far easier than displacing them.

Person trips against vehicle trips

For anything larger than a small scheme, build the assessment from person trips and then apply mode share, rather than pulling vehicle rates straight from the database.

The reason is that a vehicle trip rate from a comparator site embeds that site's mode share. If the comparator has a 70 per cent car driver share and your site has better bus provision, you have imported the wrong answer and the only remedy left is an unexplained discount.

The multi-modal approach runs in three steps. Take total person trips from multi-modal surveys for the land use. Apply a mode share drawn from evidence for your location, with census travel to work data for the relevant output areas as the usual starting point, adjusted for the accessibility of the site. Then derive vehicle trips using an occupancy assumption you state.

Illustrative example: a 180 dwelling scheme where vehicle rates taken directly from suburban comparators produced around 105 arrivals in the morning peak. Rebuilding from person trips with a mode share reflecting a nearby rail station and bus corridor produced a figure roughly 15 per cent lower, and the authority accepted it because each step was sourced. The figures are illustrative and used to show the method.

Set the travel plan targets against the same mode share. A Travel Plan that promises a car driver share the trip generation never assumed is an easy target for an objector.

Netting off an existing use

Where the site has an existing lawful use, its trips can usually be credited against the proposal, and on a redevelopment that credit often decides whether mitigation is needed.

Three conditions tend to apply. The existing use must be lawful, which is a planning question rather than a transport one. It should have been in actual use recently enough for the trips to be real, and a building vacant for several years invites an argument that the baseline is the empty building. And the trips claimed should be evidenced, ideally by a survey of the existing use rather than by a database rate for a category the building may not match.

Surveying the existing operation before it closes is the strongest evidence available and costs far less than the argument that follows without it. Where the use has already stopped, a database rate applied to the lawful floorspace with the assumptions stated is the fallback.

Councils differ on how much credit they allow for a long vacant building, so raise it at scoping rather than in the assessment.

Peak hours and the neutral survey period

Two peaks matter and they are not always the same. The network peak is when the surrounding roads are busiest, typically the weekday morning and evening commuter hours. The development peak is when the proposal generates most trips, which for a school, a retail unit or a health centre can be the middle of the day or a Saturday.

Assess both. An assessment covering only the network peaks misses the Saturday peak of a retail or leisure use, and a highway authority will ask for it.

Survey period conventions matter for the comparator sites and for any local counts you commission. Use neutral months, avoiding August and December, and avoid school holiday weeks, because term time traffic is what the network carries for most of the year. Record the survey dates in the chapter. The same convention applies to junction turning counts and queue surveys, and mixing a holiday count with term time trip rates produces a model nobody can defend.

Presenting the rates for scrutiny

The chapter should let a reviewer reproduce the rate without asking for anything. Include the category and its description, every filter with its value, the count of sites before and after each filter, the list of retained sites with their size parameters and survey dates, the rates with mean and range, and the sensitivity test.

Councils normally expect a Transport Assessment from a suitably qualified transport planner, and professional review is strongly advised because the selection judgements are where the assessment can be challenged. The habit of documenting a data enquiry so it can be checked is the same one set out in our note on the sewer capacity enquiry in a foul sewage assessment.

And state the assumptions you are least comfortable with. An assessment that identifies its own weakest input reads as competent work, while one that presents every figure as settled invites the officer to find the soft spot themselves.

This is general guidance, not legal advice, and trip generation practice varies between highway authorities, so agree the method with yours and take advice from a qualified transport planner before relying on it. PlanScope drafts a Transport Assessment structured to UK planning policy, with the survey data and modelling outputs flagged for your consultant to supply. Browse all 66 PlanScope report types.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.